{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:BHVSO6KWJQ2ZBZMBHP5EMHUKO3","short_pith_number":"pith:BHVSO6KW","canonical_record":{"source":{"id":"2312.04350","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-07T15:12:12Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"04eb62f1659f2a785663dc68c2e84eb3245a6fdda136bc9802619a97dd90ee3a","abstract_canon_sha256":"141ab001ab2289ed9b03b4ada5d5127b54520b7b119b07d167fb35eef1661eab"},"schema_version":"1.0"},"canonical_sha256":"09eb2779564c3590e5813bfa461e8a76c725b7dff62d3a08c6fe3bf37b01ca38","source":{"kind":"arxiv","id":"2312.04350","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.04350","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"arxiv_version","alias_value":"2312.04350v3","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.04350","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"pith_short_12","alias_value":"BHVSO6KWJQ2Z","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"pith_short_16","alias_value":"BHVSO6KWJQ2ZBZMB","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"pith_short_8","alias_value":"BHVSO6KW","created_at":"2026-07-05T07:34:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:BHVSO6KWJQ2ZBZMBHP5EMHUKO3","target":"record","payload":{"canonical_record":{"source":{"id":"2312.04350","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-07T15:12:12Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"04eb62f1659f2a785663dc68c2e84eb3245a6fdda136bc9802619a97dd90ee3a","abstract_canon_sha256":"141ab001ab2289ed9b03b4ada5d5127b54520b7b119b07d167fb35eef1661eab"},"schema_version":"1.0"},"canonical_sha256":"09eb2779564c3590e5813bfa461e8a76c725b7dff62d3a08c6fe3bf37b01ca38","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:34:36.480264Z","signature_b64":"dI9JfIUBdNHplqdVaE4YUJRSsjtGJ9+JqwY//7y8BRIvvv2vwkWXs+Afjc++KUvRFfMZbZ4My2R7DXNNXjQyAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"09eb2779564c3590e5813bfa461e8a76c725b7dff62d3a08c6fe3bf37b01ca38","last_reissued_at":"2026-07-05T07:34:36.479819Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:34:36.479819Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.04350","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:34:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5/AnYllIImZZXVHULzQeCUJnemJAU6GWhAJ45g27GE5yJVIZObk52icjePePIPnCxYWYGw6UvBwtvWN5wm6yCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:23:12.016657Z"},"content_sha256":"6444deac943e73027112cee2d2aef5ef5e957187fb325a5728d338be12abd347","schema_version":"1.0","event_id":"sha256:6444deac943e73027112cee2d2aef5ef5e957187fb325a5728d338be12abd347"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:BHVSO6KWJQ2ZBZMBHP5EMHUKO3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CLadder: Assessing Causal Reasoning in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bernhard Sch\\\"olkopf, Felix Leeb, Fernando Gonzalez Adauto, Kevin Blin, Luigi Gresele, Max Kleiman-Weiner, Mrinmaya Sachan, Ojasv Kamal, Yuen Chen, Zhiheng Lyu, Zhijing Jin","submitted_at":"2023-12-07T15:12:12Z","abstract_excerpt":"The ability to perform causal reasoning is widely considered a core feature of intelligence. In this work, we investigate whether large language models (LLMs) can coherently reason about causality. Much of the existing work in natural language processing (NLP) focuses on evaluating commonsense causal reasoning in LLMs, thus failing to assess whether a model can perform causal inference in accordance with a set of well-defined formal rules. To address this, we propose a new NLP task, causal inference in natural language, inspired by the \"causal inference engine\" postulated by Judea Pearl et al."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.04350","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2312.04350/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:34:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GjcZCieL1YfseZs0iKigXRTSNZIKJyeHPy42S0iMntNB6cVE8/19KisQZkBfPHgesyuvNxfW94BfQdKAYaopDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:23:12.017568Z"},"content_sha256":"32115518ab2df824f9049b5bfe6466aa7d932ba49403301a4150fd9feb5fdcf1","schema_version":"1.0","event_id":"sha256:32115518ab2df824f9049b5bfe6466aa7d932ba49403301a4150fd9feb5fdcf1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BHVSO6KWJQ2ZBZMBHP5EMHUKO3/bundle.json","state_url":"https://pith.science/pith/BHVSO6KWJQ2ZBZMBHP5EMHUKO3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BHVSO6KWJQ2ZBZMBHP5EMHUKO3/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-09T00:23:12Z","links":{"resolver":"https://pith.science/pith/BHVSO6KWJQ2ZBZMBHP5EMHUKO3","bundle":"https://pith.science/pith/BHVSO6KWJQ2ZBZMBHP5EMHUKO3/bundle.json","state":"https://pith.science/pith/BHVSO6KWJQ2ZBZMBHP5EMHUKO3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BHVSO6KWJQ2ZBZMBHP5EMHUKO3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:BHVSO6KWJQ2ZBZMBHP5EMHUKO3","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"141ab001ab2289ed9b03b4ada5d5127b54520b7b119b07d167fb35eef1661eab","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-07T15:12:12Z","title_canon_sha256":"04eb62f1659f2a785663dc68c2e84eb3245a6fdda136bc9802619a97dd90ee3a"},"schema_version":"1.0","source":{"id":"2312.04350","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.04350","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"arxiv_version","alias_value":"2312.04350v3","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.04350","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"pith_short_12","alias_value":"BHVSO6KWJQ2Z","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"pith_short_16","alias_value":"BHVSO6KWJQ2ZBZMB","created_at":"2026-07-05T07:34:36Z"},{"alias_kind":"pith_short_8","alias_value":"BHVSO6KW","created_at":"2026-07-05T07:34:36Z"}],"graph_snapshots":[{"event_id":"sha256:32115518ab2df824f9049b5bfe6466aa7d932ba49403301a4150fd9feb5fdcf1","target":"graph","created_at":"2026-07-05T07:34:36Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2312.04350/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The ability to perform causal reasoning is widely considered a core feature of intelligence. In this work, we investigate whether large language models (LLMs) can coherently reason about causality. Much of the existing work in natural language processing (NLP) focuses on evaluating commonsense causal reasoning in LLMs, thus failing to assess whether a model can perform causal inference in accordance with a set of well-defined formal rules. To address this, we propose a new NLP task, causal inference in natural language, inspired by the \"causal inference engine\" postulated by Judea Pearl et al.","authors_text":"Bernhard Sch\\\"olkopf, Felix Leeb, Fernando Gonzalez Adauto, Kevin Blin, Luigi Gresele, Max Kleiman-Weiner, Mrinmaya Sachan, Ojasv Kamal, Yuen Chen, Zhiheng Lyu, Zhijing Jin","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-07T15:12:12Z","title":"CLadder: Assessing Causal Reasoning in Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.04350","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6444deac943e73027112cee2d2aef5ef5e957187fb325a5728d338be12abd347","target":"record","created_at":"2026-07-05T07:34:36Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"141ab001ab2289ed9b03b4ada5d5127b54520b7b119b07d167fb35eef1661eab","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-07T15:12:12Z","title_canon_sha256":"04eb62f1659f2a785663dc68c2e84eb3245a6fdda136bc9802619a97dd90ee3a"},"schema_version":"1.0","source":{"id":"2312.04350","kind":"arxiv","version":3}},"canonical_sha256":"09eb2779564c3590e5813bfa461e8a76c725b7dff62d3a08c6fe3bf37b01ca38","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"09eb2779564c3590e5813bfa461e8a76c725b7dff62d3a08c6fe3bf37b01ca38","first_computed_at":"2026-07-05T07:34:36.479819Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:34:36.479819Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dI9JfIUBdNHplqdVaE4YUJRSsjtGJ9+JqwY//7y8BRIvvv2vwkWXs+Afjc++KUvRFfMZbZ4My2R7DXNNXjQyAA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:34:36.480264Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.04350","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6444deac943e73027112cee2d2aef5ef5e957187fb325a5728d338be12abd347","sha256:32115518ab2df824f9049b5bfe6466aa7d932ba49403301a4150fd9feb5fdcf1"],"state_sha256":"dd01d505995c7603d1e9cbca75b4e42b08b7448bf5cfaa968644eca0a24d1688"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Swk6DXPGMWbN0OBzC7AXbDuBu3qB6WAns/wrchVcU47AUw2JYAoTB9MOxG8jkS1Kr05VspT/l15Qq+0f6dQaCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T00:23:12.024593Z","bundle_sha256":"26e56fb27a7318eaa2d05d23b070435a20b08ba0f176505a7dc78b305f4f73ae"}}